CN104573054A - Information pushing method and equipment - Google Patents
Information pushing method and equipment Download PDFInfo
- Publication number
- CN104573054A CN104573054A CN201510030145.0A CN201510030145A CN104573054A CN 104573054 A CN104573054 A CN 104573054A CN 201510030145 A CN201510030145 A CN 201510030145A CN 104573054 A CN104573054 A CN 104573054A
- Authority
- CN
- China
- Prior art keywords
- media event
- user
- event
- news information
- news
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 206010028916 Neologism Diseases 0.000 claims description 8
- 238000012552 review Methods 0.000 claims description 4
- 238000011161 development Methods 0.000 description 6
- 230000018109 developmental process Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005303 weighing Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008595 infiltration Effects 0.000 description 2
- 238000001764 infiltration Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides an information pushing method. The method comprises the following steps: determining the pushing degree of a user by each news event according to the pre-obtained attention to each news event by the user and the correlation degree between the news events; determining the news events needing to be pushed to the user according to the pushing degree; and pushing news information in the news events needing to be pushed to the user to the user. The attention to each news event by the user and the correlation degree between the news events are obtained and the condition of pushing the news information conforming to the interests of the user to the user becomes possible so that the accuracy of pushing the news information meeting the user requirements is remarkably improved and better experiences are brought to the user. Furthermore, the embodiment of the invention provides information pushing equipment.
Description
Technical field
Embodiments of the present invention relate to Internet information technique process field, and more specifically, embodiments of the present invention relate to a kind of information-pushing method and equipment.
Background technology
This part embodiments of the present invention be intended to for stating in claims provide background or context.Description is not herein because be included in just admit it is prior art in this part.
At present, along with the explosivity day by day of quantity of information increases, each large door news website or application are all proposed the mode pushing news information to user.
A kind of conventional mode be the classification liked according to user carry out news information push such as, amusement, or more segment large class is determined group, such as, amusement-Eight Diagrams, etc.
Summary of the invention
But user may be only interested in some news information in some classifications, is not all interested in all news informations in this classification under normal circumstances.
Inevitably, conventionally push news information to user according to classification, may will make to push result and can not meet user's requirement, also therefore, if user needs to see oneself interested news information, just needs triggering is more could realize with news website or the mutual of news application.
For this reason, be starved of a kind of method of information pushing of improvement, become possibility to make pushing to user the news information more meeting its reading requirement and interest, promote the degree of accuracy that pushed information meets user's requirement.
In the present context, embodiments of the present invention are expected to provide a kind of information-pushing method and information pushing equipment.
In the first aspect of embodiment of the present invention, provide a kind of information-pushing method, comprising: according to the user obtained in advance to the attention rate of each media event, and, correlation degree between each media event, determines the propelling movement degree of each media event to described user; The media event needing to push to described user is determined according to described propelling movement degree; News information in the media event needing to push to described user is pushed to described user.
In the second aspect of embodiment of the present invention, provide a kind of information pushing equipment, comprise: the first determination module, be configured for according to the user obtained in advance the attention rate of each media event, with, correlation degree between each media event, determines the propelling movement degree of each media event to described user; Second determination module, is configured for and determines according to described propelling movement degree the media event that needs push to described user; Pushing module, is configured for and the news information in the media event needing to push to described user is pushed to described user.
According to information-pushing method and the information pushing equipment of embodiment of the present invention, can by user to the incidence relation between the attention rate of each media event and each media event, thus calculate the propelling movement degree of each media event for user, and the news information in media event high for propelling movement degree is pushed to user, become possibility to make pushing to user the news information more meeting user's reading requirement and interest, thus improve the degree of accuracy pushing and meet the news information that user requires.
summary of the invention
The present inventor finds, prior art is because only distinguished the classification of user preferences, and make such division too coarse, because user may be only interested in some news information in some classifications, be not all interested in all news informations in this classification under normal circumstances.Therefore, such as, if according to user at ordinary times to the concern situation of media event, " amusement " class media event whether can be paid close attention to more more, and in conjunction with the association situation between each media event, determined and whether push certain media event to user.So just more can meet the news information of its interest according to the feature of user preferences and media event to user's propelling movement simultaneously.
After describing ultimate principle of the present invention, lower mask body introduces various non-limiting embodiment of the present invention.
application scenarios overview
First with reference to figure 1, user can browse various news information in internet device 101, this internet device 101 can be desktop computer or hand-held smart machine etc., as long as this internet device 101 can pass through internet connection server 102, the news information that just can receive server 102 propelling movement is browsed for user.And internet device 101 can also be installed various news application, such as " Netease's news " etc., user also can come mutual with server 102 by news application and browse various news information.
illustrative methods
Below in conjunction with the application scenarios of Fig. 1, be described with reference to Figure 2 the method for information pushing according to exemplary embodiment of the invention.It should be noted that above-mentioned application scenarios is only that embodiments of the present invention are unrestricted in this regard for the ease of understanding spirit of the present invention and principle and illustrating.On the contrary, embodiments of the present invention can be applied to applicable any scene.
Step 201: according to the user obtained in advance to the attention rate of each media event, and the correlation degree between each media event, determines the propelling movement degree of each media event to described user.
In the present embodiment, media event refers to the focus incident that has the multiple news informations be associated, such as " the large wedding of star XX ", so may have under this media event multiple " star XX tries wedding gauze kerchief on ", " star XX determines in such a month, and on such a day to have a wedding " etc. multiple be associated and all related with this media event news information.Wherein, news information can be one section of news article belonging to certain media event, such as " star XX tries wedding gauze kerchief on ".
In the present embodiment, the correlation degree obtained in step 201 between each media event comprises: the incidence relation obtaining each media event, and calculates the correlation degree between each relevant media event.Wherein, this correlation degree can indicate the incidence relation of each media event and other media events is near or far away.The incidence relation of media event can comprise successor relationship and side shoot relation, wherein, successor relationship can be: a media event is identical with the principal dimensions of another media event and time of origin is later than another media event, and described principal dimensions refers to multiple dimensions (such as 2 dimensions: content and personage) of default energy presentation of events key message; Described side shoot relation can be: a media event is identical with the partial dimensional of another media event, and such as, the content of a media event is identical from another media event and personage is different.
Correlation degree between each media event that described calculating is relevant, comprising: determine the associated weights value between each media event according to the incident distance between described each relevant media event and event hot value; Described incident distance can comprise: the time gap between each media event described and characteristic distance.
In the process determining associated weights value, the successor of each media event or the mapping model of side shoot event can be built according to the flow of event structure calculated, be called event evolves correlation model.Then according to the incident distance (comprising time gap and characteristic distance) between event and event hot value, the limit weight of the oriented association of each event can be given as associated weights value, the event A such as associated and the time gap of B or characteristic distance nearer, namely, the time that A and B occurs is nearer, the distance of feature space is nearer, then the associated weights value between A and B is larger.
Wherein, the determination mode of time gap can comprise steps A 1 ~ steps A 2 as follows:
Steps A 1: obtain the time of origin relation between the event hot value of each media event and each media event from the flow of event structure set up in advance.
In the present embodiment, can set up flow of event structure in advance, the relation of this flow of event representation on the time line between each media event, wherein, each media event all comprises the event hot value of the popular degree representing oneself.This event hot value can obtain by the user operation situation of user to media event being weighted, wherein, user operation situation can comprise: the quantity of news information identical in the click volume of user, user's clicking rate and same media event, etc.
When setting up flow of event structure, the mode of Down-Up hierarchical clustering can be adopted, when two media events eigencenter very close to time, if keyword feature is very near on principal dimensions, (personage related in such as media event is identical with content, other dimensions there occurs change), and the time continues and develops backward, so think that this is the media events of two continuity development, time posterior media event is the successor of previous media event; And if keyword feature is less than a less threshold value in the distance of part principal dimensions, and (this threshold value can arrange empirical value by those skilled in the art, such as, distance is less than 0.2), and in some secondary dimension distance more than a no small threshold value (such as distance is more than 0.7), namely, figure picture in media event is same, and content change and the time be also continuity, so the posterior media event of time of origin is a side shoot event (may be such as some focus comments to certain media event) of last media event.Therefore, flow of event structure can indicate each media event structure on the time line, thus can obtain the Time evolution circuit of each flash-news event, and relevant side shoot event.
Steps A 2: determine the time gap between each media event according to time of origin relation.
Can determine that the time gap between each media event is how many according to time of origin relation again.
Wherein, in the present embodiment, when pushing news information to certain user, can obtain the attention rate of this user to each media event in advance, it is large or little that this attention rate can indicate user to the interested degree of each media event.Obtain the attention rate of user to each media event in step 201, specifically can comprise: according to the news information that user is browsed in preset time period, obtain user to the attention rate of media event.This step, when specific implementation, can comprise step B1 as follows ~ step B2:
Step B1: obtain user in Preset Time browsed headline and/or body browse news information.
Wherein, user can by adding up the media event of this user in the recent period belonging to the news information effectively browsed such as (such as two months or three months) to the attention rate of each media event, such as, can according to user to browsing time of news information in media event, browse quantity and whether behavior weighting commented on etc. to news information and obtain.
When obtaining the attention rate of user to each media event, user's attention rate model can be trained.Specifically when training user attention rate model, first can get user in the past in Preset Time, such as, in three months, browsed headline or body, or simultaneously browsed headline and body browse news information.
Step B2: determine the attention rate of user to media event according to the described news features having browsed news information; Wherein, described news features specifically can comprise: user browses to the review information of described news information, user the classification that browsing time of described news information and user browse news information.
Then according to user to review information and the browsing time of browsing news information, and user browses the news category of news information, three is weighted and obtains user's attention rate thus train user's attention rate model, then can determine the attention rate of certain user to certain media event according to the news features of browsing information.
Concrete, the flow of event structure in steps A 1 is built by step C1 shown below ~ step C4:
Step C1: determine the feature sequence of terms included by news information.
First, determine the feature sequence of terms included by news information, participle and part-of-speech tagging can be carried out to news information, obtain initial sequence of terms, and delete in initial sequence of terms and influential word is not had to news features, to obtain feature sequence of terms.General text segmenter can be adopted to carry out participle and part-of-speech tagging to the title of news information and text when implementing, and analysis and filter falls the unconspicuous word of feature in the result of part-of-speech tagging, such as auxiliary word, pronoun, conjunction, preposition and modal particle etc., eliminate so more accurately and do not have influential word to news features.Wherein, carry out the dictionary for word segmentation that participle adopts in present embodiment and can comprise neologisms.Neologisms test and monitoring algorithm can be adopted in actual applications to find out the significant neologisms occurred in news information, then add in dictionary for word segmentation in time.
Step C2: according to the word weighted value of each feature word in described feature sequence of terms, is expressed as corresponding news documents vector sum keyword vector by each news information.
In feature sequence of terms, each feature word has a word weighted value, according to the word weighted value of each feature word, each news information is expressed as corresponding news documents vector sum keyword vector.
Concrete, described step C2 can comprise step D1 as follows ~ step D5:
Step D1: according to the word feature of the feature word in described feature sequence of terms, calculates the word weighted value of each feature word; Wherein, institute's predicate feature comprises: whether the temperature of part of speech, word frequency, word, feature word are neologisms or proprietary word and the position of word in news information.
First, the word weighted value of each feature word is calculated according to the word feature of each feature word, wherein, word feature comprises: whether the temperature of the part of speech of feature word, word frequency, word, feature word are neologisms or proprietary word and the position of word in news information, etc.Wherein, proprietary word can be the proper noun adopting named entity recognition method to detect from news information, obtains name, place name, exabyte or date etc. as detected.Named entity recognition method can simply mate from rule base, also can adopt the method identification of more complicated machine learning, such as, adopt condition random field algorithm to build model of cognition lamp.Word frequency is the frequency that a feature word occurs in certain news information.Then can calculate the weight of each feature word in news information according to word feature, generally the set of the feature word that weight is high often can represent the meaning of this section of news information.
Wherein, word weighing computation method is a lot, can being simply such as the weighting based on TF-IDF (termfrequency – inverse document frequency, word frequency and inverse document frequency), obtaining weighted value as being directly weighted according to word feature.In addition, the more accurate weighing computation method based on TextRank can be adopted, in order to accurate Calculation weight more, can in TextRank word network, the limit of each word is arranged to the limit weight in word network, the limit weight of the word that such as position is far away in word network is lower, and the limit weight of word in word earlier in word network or title is higher.
Step D2: feature word word weighted value being exceeded default weight threshold is expressed as the sparse vector pattern of band word weighted value, to obtain news documents vector.
After obtaining word weighted value, a part of Feature Words of word weighted value more than a default weight threshold (such as 0.4) can be got, be expressed as the sparse vector pattern of cum rights weight values thus obtain news documents vector.Such as: " iPhone ": 0.91, " apple ": 0.82, " issue ": 0.54 ....
Step D3: judging characteristic word, whether in keyword dictionary, if so, then enters step D4.
A keyword dictionary can be safeguarded in the present embodiment.Specifically according to the temperature of word in the word frequency in dictionary for word segmentation and neologisms dictionary, can obtain some weight ratios comparatively large (being such as greater than 0.4) in conjunction with part of speech and named entity recognition algorithm and a keyword dictionary of the implication of media event can be represented.This keyword dictionary can dynamically update, and can comprise the large class such as personage, time, place, event, the result of the named entity recognition often of personage wherein, such as " Apple ", " google " etc.
Step D4: the keyword that word weighted value exceedes default weight threshold is combined as keyword set.
If feature word is the keyword in keyword dictionary, and weight exceedes certain threshold value (such as more than 0.5) just puts it in keyword set.
Step D5: the sparse vector pattern keyword in described keyword set being expressed as band word weighted value, to obtain keyword vector.
Then be also expressed as sparse vector to the word in keyword set, and it can be used as keyword vector, this keyword vector may be used for the coupling which classification is follow-up media event be classified as.
Further, after news information is expressed as corresponding news documents vector sum keyword vector, duplicate removal (namely removing the news information of repetition) can be carried out to new news information, and carry out duplicate removal with history news information, then classification is re-started to the news information after duplicate removal, such as, be classified as each large class such as " internet ", " amusement " or " important news ".
When carrying out news information duplicate removal, distance between news documents vector is less than certain threshold value (such as, be less than 0.1) news information be defined as same section news information, namely both similarities are greater than a certain threshold value and then think that these two sections of news informations are same.Wherein, the calculating formula of similarity calculating news information adopts COS distance, as shown in formula ():
Wherein, S represents the similarity of news information A and B, and the length of news documents vector is sparse n dimension, and n gets the quantity of participle in dictionary for word segmentation usually, and this dictionary for word segmentation is the dictionary for word segmentation obtained in step D1; The value of the more similar then S of A with B is more close to 1, more dissimilar more close to 0.
Wherein, can be obtained the training of markd news information sample by machine learning method the disaggregated model of classifying after news information duplicate removal, such as using news documents vector as feature, by the news information of the good classification of handmarking as training sample, build training aids by the method for support vector machine (SVM), then new news information is classified.Wherein, the object of classifying to news information is the accurate identification being convenient to follow-up media event, because the word weighted value of the keyword of different classes of news information is different, and the sub-category accuracy that can improve media event identification.
After having introduced and how obtaining news documents vector sum keyword vector, then enter step C3: according to described news documents vector sum keyword vector, for described news information sets up corresponding event base.
After the news documents vector sum keyword vector obtaining news information, for news information sets up corresponding event base.Wherein, described step C3 specifically can comprise: first judge whether a news information belongs to existing media event in event base, if, then described news information is classified as existing media event, if not, then set up new media event for described news information, and described new media event is saved in event base.
In the present embodiment, existing media event cluster basis compares new news information, if new media event is in the clustering of old media event, then think that this new media event belongs to old media event.If the extension of (such as the dimension of some high priest and content is constant) on the certain orientation that new news documents vector (combining keyword vector) clusters in certain old media event, so then determines that new media event is the infiltration and development of old media event.Such as news first " iPhone 6Plus deposits new quality problems and probably recalls on a large scale " and news second " apple iPhone 6Plus problem exist already but recall rumor wrong ", the content of these two sections of news informations is " iPhone6Plus has problems and recalls rumor ", therefore be consistent on the main direction, but news second is issued, so think that news second is the infiltration and development of news first for slower than news first one day.
And for example fruit, the Main way (such as personage is constant) that this new media event clusters is identical with old media event, and it is large to cluster, and distance is near, so just thinks the newly-developed event of old media event.Such as early stage news information be " star's first probably contract company B ", and the news information after several days reports " star's first contracted the third company ", then the news information can thinking below is the recent development event about star's first earliest events.If new media event clusters and leaves existing media event and cluster all more than a larger threshold value, so just set up new events for new media event, by news documents vector and the vectorial proper vector as this New News event of keyword of this media event.Wherein, newer media event be whether new events or development event time, also consider time factor, if think same media event then its time of origin want consistent, if the time of origin of the then new media event of follow-up developments event must be later than the time of origin of old media event.
Step C4: the time of origin of the media event in described event base according to each media event is sorted, and the event hot value calculating each media event according to affair character.
After foundation or upgrading event base, each media event in event base is sorted according to the time of origin of each media event again, and calculate the event hot value of each media event according to affair character.Wherein, affair character can comprise: user is to the number of the clicking rate of the news information in media event, click volume and identical news information in a media event.The mode of weighting can be adopted to calculate event hot value.
After having introduced and how obtaining event hot value and time of origin relation from event flow structure, the time gap between each media event can be determined according to time of origin relation.
In the present embodiment, get user to the attention rate of media event and associated weights value after, calculate the propelling movement degree of each media event relative to user.Collaborative filtering can be adopted to calculate weight, wherein, user can be expressed as a weighing vector to the attention rate of media event, such as { " Hangzhou marathon ": 0.62, " volume good fortune is engaged ": 0.51,, and each media event has the corresponding mapping event be associated, the concrete mode so calculating propelling movement degree represents as shown in formula (two):
Wherein, R
irepresent the recommendation degree of media event i for this user; u
jfor this user is to the attention rate of media event j; w
ijfor the weight on the jth bar media event limit of the association of media event i.Wherein, the positive integer of i and j all for being greater than 0.
After having introduced the process how calculating propelling movement degree, return Fig. 2, enter step 202: determine the media event needing to push to described user according to described propelling movement degree.
According to the propelling movement degree of each media event relative to user, determine which media event needs to recommend user.Such as, the propelling movement degree news information be greater than in the media event of 0.6 needs to be pushed to user, then the value according to the propelling movement degree of each media event filters out ineligible media event.
Step 203: the news information in the media event needing to push to described user is pushed to described user.
So, when needing have new news information to issue in the media event pushed to this user, just new news information can be pushed to user, so that user conveniently browses to its interested news information.Be understandable that, in actual applications, first can judge that whether new news information user is browsed, just do not need repetition to push to user if browsed, and if be not browsedly pushed to user again.
Optionally, can also comprise before step 203:
Step 200: with reference to source and the quality of described news information, carries out quality-ordered in the media event of the news information under each media event belonging to it.
Be understandable that, because have multiple news information in a media event, then can to reference to the source of described news information and quality, carry out quality-ordered in the media event of the news information under each media event belonging to it.Wherein, the source of news information refers to the media of the information of releasing news, and whether are such as authoritative media etc., the news information of authoritative media releasing can arrange larger weight; The quality of news information then considers the following factor, but be not limited only to these: the source of multimedia (as image, audio or video etc.), quantity and position in the form of news information and typesetting, news category, headline and content normality, news information, and, whether with marketing advertisement etc. in news information.
After quality-ordered is carried out to the news information in media event, the representative news information of the good news information of quality as the media event belonging to it can be filtered out.Doing so also can the second-rate news information of elimination, and the different news informations that simultaneously it also avoid identical media event can recommend the situation of same user.
When sorting to the news information in media event, step 203 is specifically as follows: first, obtain propelling movement degree and be greater than the media event preset and push threshold value, then news information quality-ordered in described media event being met preset quality requirement (such as quality is greater than 0.7) is pushed to described user.
Visible in the application's embodiment, can by user to the incidence relation between the attention rate of each media event and each media event, thus calculate the propelling movement degree of each media event for user, and the news information in media event high for propelling movement degree is pushed to user, become possibility to make pushing to user the news information more meeting user's reading requirement and interest, thus improve the degree of accuracy pushing and meet the news information that user requires.
example devices
After the method describing exemplary embodiment of the invention, next, with reference to figure 3 pairs of exemplary embodiment of the invention, for the equipment of information pushing,
First determination module 301, be configured for according to the user obtained in advance to the attention rate of each media event, and the correlation degree between each media event, determines the propelling movement degree of each media event to described user.
Wherein, when described first determination module 301 obtains the correlation degree between each media event, specifically can obtain the incidence relation of each media event, and the correlation degree calculated between each relevant media event, the incidence relation of described media event comprises successor relationship and side shoot relation, described successor relationship is: a media event is identical with the principal dimensions of another media event and time of origin is later than another media event, and described principal dimensions is multiple dimensions of default energy presentation of events key message; Described side shoot closes: a media event is identical with the partial dimensional of another media event.
Wherein, calculate the correlation degree between each relevant media event, comprising: determine the associated weights value between each media event according to the incident distance between described each relevant media event and event hot value; Described incident distance comprises: the time gap between each media event described and characteristic distance.
Wherein, described time gap obtains in the following manner: from the flow of event structure set up in advance, obtain the time of origin relation between the event hot value of each media event and each media event; The time gap between each media event is determined according to described time of origin relation.
Wherein, the attention rate of described user to each media event obtains in the following manner: according to the news information that user is browsed in preset time period, obtains user to the attention rate of media event.
Wherein, described according to the news information browsed in preset time period of user, obtain user to the attention rate of media event, comprising: obtain user in Preset Time browsed headline and/or body browse news information; The attention rate of user to media event is determined according to the described news features having browsed news information; Wherein, described news features comprises: user browses to the review information of described news information, user the classification that browsing time of described news information and user browse news information.
Wherein, described flow of event structure can build in the following manner:
Determine the feature sequence of terms included by news information; According to the word weighted value of each feature word in described feature sequence of terms, each news information is expressed as corresponding news documents vector sum keyword vector; According to described news documents vector sum keyword vector, for described news information sets up corresponding event base; The time of origin of media event in described event base according to each media event is sorted, and calculates the event hot value of each media event according to affair character; Described affair character comprises: user is to the number of the clicking rate of the news information in media event, click volume and identical news information in a media event.
Wherein, the described feature sequence of terms determined included by news information, can comprise: carry out participle and part-of-speech tagging to news information sample, obtain initial sequence of terms; Delete in described initial sequence of terms and influential word is not had to news features, to obtain feature sequence of terms.
Wherein, the described word weighted value according to each feature word in described feature sequence of terms, each news information is expressed as corresponding news documents vector sum keyword vector, comprise: according to the word feature of the feature word in described feature sequence of terms, calculate the word weighted value of each feature word; Wherein, institute's predicate feature comprises: whether the temperature of part of speech, word frequency, word, feature word are the position in neologisms or proprietary word and word place news information; Feature word word weighted value being exceeded default weight threshold is expressed as the sparse vector pattern of band word weighted value, to obtain news documents vector; Whether judging characteristic word is in keyword dictionary, and if so, then keyword word weighted value being exceeded default weight threshold is combined as keyword set; Keyword in described keyword set is expressed as the sparse vector pattern of band word weighted value, to obtain keyword vector.
Wherein, described according to described document vector sum keyword vector, for described news information sets up corresponding event base, can comprise: judge whether described news information belongs to existing media event in event base, if so, then described news information is classified as existing media event, if not, then set up new media event for described news information, and described new media event is saved in event base.
Second determination module 302, is configured for and determines according to described propelling movement degree the media event that needs push to described user.
Pushing module 303, is configured for and the news information in the media event needing to push to described user is pushed to described user.
Optionally, can also comprise: order module, be configured for the source with reference to described news information and quality, carry out quality-ordered in the media event of the news information under each media event belonging to it.Then corresponding, described pushing module 303 specifically can comprise: obtain submodule, is configured for the media event obtaining propelling movement degree and be greater than default propelling movement threshold value; With, push submodule, be configured for news information quality-ordered in described media event being met preset quality requirement and be pushed to described user.
In the application's embodiment, can by user to the incidence relation between the attention rate of each media event and each media event, thus calculate the propelling movement degree of each media event for user, and the news information in media event high for propelling movement degree is pushed to user, become possibility to make pushing to user the news information more meeting user's reading requirement and interest, thus improve the degree of accuracy pushing and meet the news information that user requires.
Although it should be noted that the some devices or sub-device that are referred to information pushing equipment in above-detailed, this division is only not enforceable.In fact, according to the embodiment of the present invention, the Characteristic and function of two or more devices above-described can be specialized in one apparatus.Otherwise, the Characteristic and function of an above-described device can Further Division for be specialized by multiple device.
In addition, although describe the operation of the inventive method in the accompanying drawings with particular order, this is not that requirement or hint must perform these operations according to this particular order, or must perform the result that all shown operation could realize expectation.Additionally or alternatively, some step can be omitted, multiple step be merged into a step and perform, and/or a step is decomposed into multiple step and perform.
Although describe spirit of the present invention and principle with reference to some embodiments, but should be appreciated that, the present invention is not limited to disclosed embodiment, can not combine to be benefited to the feature that the division of each side does not mean that in these aspects yet, this division is only the convenience in order to state.The present invention is intended to contain the interior included various amendment of spirit and scope and the equivalent arrangements of claims.
Although it should be noted that the some devices or sub-device that are referred to information pushing equipment in above-detailed, this division is only not enforceable.In fact, according to the embodiment of the present invention, the Characteristic and function of two or more devices above-described can be specialized in one apparatus.Otherwise, the Characteristic and function of an above-described device can Further Division for be specialized by multiple device.
In addition, although describe the operation of the inventive method in the accompanying drawings with particular order, this is not that requirement or hint must perform these operations according to this particular order, or must perform the result that all shown operation could realize expectation.Additionally or alternatively, some step can be omitted, multiple step be merged into a step and perform, and/or a step is decomposed into multiple step and perform.
Although describe spirit of the present invention and principle with reference to some embodiments, but should be appreciated that, the present invention is not limited to disclosed embodiment, can not combine to be benefited to the feature that the division of each side does not mean that in these aspects yet, this division is only the convenience in order to state.The present invention is intended to contain the interior included various amendment of spirit and scope and the equivalent arrangements of claims.
Accompanying drawing explanation
By reference to accompanying drawing reading detailed description hereafter, above-mentioned and other objects of exemplary embodiment of the invention, feature and advantage will become easy to understand.In the accompanying drawings, show some embodiments of the present invention by way of example, and not by way of limitation, wherein:
Fig. 1 schematically shows the application scenarios Organization Chart according to embodiment of the present invention;
Fig. 2 schematically shows the process flow diagram according to information-pushing method embodiment of the present invention;
Fig. 3 schematically shows the structured flowchart according to information pushing apparatus embodiments of the present invention;
In the accompanying drawings, identical or corresponding label represents identical or corresponding part.
Embodiment
Below with reference to some illustrative embodiments, principle of the present invention and spirit are described.Should be appreciated that providing these embodiments is only used to enable those skilled in the art understand better and then realize the present invention, and not limit the scope of the invention by any way.On the contrary, provide these embodiments to be to make the disclosure more thorough and complete, and the scope of the present disclosure intactly can be conveyed to those skilled in the art.
One skilled in the art will appreciate that embodiments of the present invention can be implemented as a kind of system, device, equipment, method or computer program.Therefore, the disclosure can be implemented as following form, that is: hardware, completely software (comprising firmware, resident software, microcode etc.) completely, or the form that hardware and software combines.
According to the embodiment of the present invention, a kind of method and apparatus of information pushing is proposed.
In this article, any number of elements in accompanying drawing is all unrestricted for example, and any name is all only for distinguishing, and does not have any limitation.
Below with reference to some representative embodiments of the present invention, explaination principle of the present invention and spirit in detail.
Claims (13)
1. an information-pushing method, comprising:
According to the user obtained in advance to the attention rate of each media event, and the correlation degree between each media event, determines the propelling movement degree of each media event to described user;
The media event needing to push to described user is determined according to described propelling movement degree;
News information in the media event needing to push to described user is pushed to described user.
2. method according to claim 1, the correlation degree between each media event described obtains in the following manner:
Obtain the incidence relation of each media event, and the correlation degree calculated between each relevant media event, the incidence relation of described media event comprises successor relationship and side shoot relation, described successor relationship is: a media event is identical with the principal dimensions of another media event and time of origin is later than another media event, and described principal dimensions is multiple dimensions of default energy presentation of events key message; Described side shoot closes: a media event is identical with the partial dimensional of another media event.
3. method according to claim 2, the correlation degree between each media event that described calculating is relevant, comprising:
The associated weights value between each media event is determined according to the incident distance between described each relevant media event and event hot value; Described incident distance comprises: the time gap between each media event described and characteristic distance.
4. method according to claim 3, described time gap obtains in the following manner:
The time of origin relation between each media event is obtained from the flow of event structure set up in advance;
The time gap between each media event is determined according to described time of origin relation.
5. method according to claim 1, the attention rate of described user to each media event obtains in the following manner:
According to the news information that user is browsed in preset time period, obtain user to the attention rate of media event.
6. method according to claim 5, the described news information browsed in preset time period according to user, obtains user to the attention rate of media event, comprising:
Obtain user in Preset Time browsed headline and/or body browse news information;
The attention rate of user to media event is determined according to the described news features having browsed news information; Wherein, described news features comprises: user browses to the review information of described news information, user the classification that browsing time of described news information and user browse news information.
7. method according to claim 6, described flow of event structure builds in the following manner:
Determine the feature sequence of terms included by news information;
According to the word weighted value of each feature word in described feature sequence of terms, each news information is expressed as corresponding news documents vector sum keyword vector;
According to described news documents vector sum keyword vector, for described news information sets up corresponding event base;
The time of origin of media event in described event base according to each media event is sorted, and calculates the event hot value of each media event according to affair character; Described affair character comprises: user is to the number of the clicking rate of the news information in media event, click volume and identical news information in a media event.
8. method according to claim 7, the described feature sequence of terms determined included by news information, comprising:
Participle and part-of-speech tagging are carried out to news information sample, obtains initial sequence of terms;
Delete in described initial sequence of terms and influential word is not had to news features, to obtain feature sequence of terms.
9. method according to claim 7, the described word weighted value according to each feature word in described feature sequence of terms, is expressed as corresponding news documents vector sum keyword vector, comprises by each news information:
According to the word feature of the feature word in described feature sequence of terms, calculate the word weighted value of each feature word; Wherein, institute's predicate feature comprises: whether the temperature of part of speech, word frequency, word, feature word are the position in neologisms or proprietary word and word place news information;
Feature word word weighted value being exceeded default weight threshold is expressed as the sparse vector pattern of band word weighted value, to obtain news documents vector;
Whether judging characteristic word is in keyword dictionary, and if so, then keyword word weighted value being exceeded default weight threshold is combined as keyword set;
Keyword in described keyword set is expressed as the sparse vector pattern of band word weighted value, to obtain keyword vector.
10. method according to claim 7, described according to described document vector sum keyword vector, for described news information sets up corresponding event base, comprising:
Judge whether described news information belongs to existing media event in event base, if so, then described news information is classified as existing media event, if not, then set up new media event for described news information, and described new media event is saved in event base.
11. methods according to claim 7, also comprise:
With reference to source and the quality of described news information, carry out quality-ordered in the media event of the news information under each media event belonging to it.
12. methods according to claim 11, the news information in the media event that described needs push to described user is pushed to described user, comprising:
Obtain propelling movement degree and be greater than the media event preset and push threshold value;
Quality-ordered in obtained media event is met the news information that preset quality requires and be pushed to described user.
13. 1 kinds of information pushing equipment, comprising:
First determination module, be configured for according to the user obtained in advance to the attention rate of each media event, and the correlation degree between each media event, determines the propelling movement degree of each media event to described user;
Second determination module, is configured for and determines according to described propelling movement degree the media event that needs push to described user;
Pushing module, is configured for and the news information in the media event needing to push to described user is pushed to described user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510030145.0A CN104573054B (en) | 2015-01-21 | 2015-01-21 | A kind of information-pushing method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510030145.0A CN104573054B (en) | 2015-01-21 | 2015-01-21 | A kind of information-pushing method and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104573054A true CN104573054A (en) | 2015-04-29 |
| CN104573054B CN104573054B (en) | 2018-06-01 |
Family
ID=53089116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510030145.0A Active CN104573054B (en) | 2015-01-21 | 2015-01-21 | A kind of information-pushing method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104573054B (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105007528A (en) * | 2015-07-06 | 2015-10-28 | 无锡天脉聚源传媒科技有限公司 | Method and device for searching video |
| CN105787049A (en) * | 2016-02-26 | 2016-07-20 | 浙江大学 | Network video hotspot event finding method based on multi-source information fusion analysis |
| CN106202501A (en) * | 2016-07-20 | 2016-12-07 | 宁波公众信息产业有限公司 | A kind of information analysis system |
| CN106202563A (en) * | 2016-08-02 | 2016-12-07 | 西南石油大学 | A kind of real time correlation evental news recommends method and system |
| CN106372113A (en) * | 2016-08-22 | 2017-02-01 | 上海亿账通互联网科技有限公司 | News content pushing method and system |
| CN106407361A (en) * | 2016-09-07 | 2017-02-15 | 北京百度网讯科技有限公司 | Method and device for pushing information based on artificial intelligence |
| CN106557508A (en) * | 2015-09-28 | 2017-04-05 | 北京神州泰岳软件股份有限公司 | A kind of text key word extracting method and device |
| CN106557513A (en) * | 2015-09-29 | 2017-04-05 | 腾讯科技（深圳）有限公司 | Event information method for pushing and event information pusher |
| CN106570144A (en) * | 2016-02-05 | 2017-04-19 | 中科鼎富（北京）科技发展有限公司 | Method and apparatus for recommending information |
| CN106933878A (en) * | 2015-12-30 | 2017-07-07 | 腾讯科技（北京）有限公司 | A kind of information processing method and device |
| CN106951435A (en) * | 2017-02-08 | 2017-07-14 | 广州神马移动信息科技有限公司 | News recommends method, equipment and programmable device |
| CN106993030A (en) * | 2017-03-22 | 2017-07-28 | 北京百度网讯科技有限公司 | Information-pushing method and device based on artificial intelligence |
| CN107169012A (en) * | 2017-03-31 | 2017-09-15 | 百度在线网络技术（北京）有限公司 | POI recommends method, device, equipment and computer-readable recording medium |
| CN107239537A (en) * | 2017-06-01 | 2017-10-10 | 福建中金在线信息科技有限公司 | A kind of method of document display, device, electronic equipment and readable storage medium storing program for executing |
| CN107273355A (en) * | 2017-06-12 | 2017-10-20 | 大连理工大学 | A kind of Chinese word vector generation method based on words joint training |
| CN107402925A (en) * | 2016-05-19 | 2017-11-28 | 阿里巴巴集团控股有限公司 | Information-pushing method and device |
| CN107506367A (en) * | 2017-07-03 | 2017-12-22 | 阿里巴巴集团控股有限公司 | It is determined that the method, apparatus and server of application displaying content |
| CN107657067A (en) * | 2017-11-14 | 2018-02-02 | 国网山东省电力公司电力科学研究院 | A kind of quick method for pushing of frontier science and technology information and system based on COS distance |
| CN108241631A (en) * | 2016-12-23 | 2018-07-03 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of pushed information |
| CN108470046A (en) * | 2018-03-07 | 2018-08-31 | 中国科学院自动化研究所 | Media event sort method and system based on media event search statement |
| CN108519980A (en) * | 2018-01-31 | 2018-09-11 | 广东易联创富集团有限公司 | News push method, apparatus, platform, computer readable storage medium |
| CN108563630A (en) * | 2018-03-21 | 2018-09-21 | 上海蔚界信息科技有限公司 | A kind of construction method of text analyzing knowledge base |
| CN108694211A (en) * | 2017-04-11 | 2018-10-23 | 腾讯科技（深圳）有限公司 | Using distribution method and device |
| CN109241277A (en) * | 2018-07-18 | 2019-01-18 | 北京航天云路有限公司 | The method and system of text vector weighting based on news keyword |
| CN110110303A (en) * | 2019-03-28 | 2019-08-09 | 苏州八叉树智能科技有限公司 | Newsletter archive generation method, device, electronic equipment and computer-readable medium |
| CN110335654A (en) * | 2019-07-03 | 2019-10-15 | 重庆邮电大学 | A kind of information extraction method of electronic health record, system and computer equipment |
| CN110837609A (en) * | 2019-11-12 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Social information processing method and related product |
| CN110990690A (en) * | 2019-11-12 | 2020-04-10 | 上海易点时空网络有限公司 | Post recommendation method and device |
| CN111460289A (en) * | 2020-03-27 | 2020-07-28 | 北京百度网讯科技有限公司 | News information pushing method and device |
| CN111966784A (en) * | 2020-07-14 | 2020-11-20 | 深圳中泓在线股份有限公司 | Information recommendation method, electronic device and storage medium |
| CN112613317A (en) * | 2020-12-30 | 2021-04-06 | 中国农业银行股份有限公司 | Text data cleaning method and device |
| CN114564675A (en) * | 2022-04-28 | 2022-05-31 | 深圳格隆汇信息科技有限公司 | Information recommendation method and device and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070174167A1 (en) * | 2005-05-20 | 2007-07-26 | Stefano Natella | Derivative relationship news event reporting |
| CN101174273A (en) * | 2007-12-04 | 2008-05-07 | 清华大学 | News event detecting method based on metadata analysis |
| CN102073631A (en) * | 2009-11-19 | 2011-05-25 | 凌坚 | Video news unit dividing method by using association rule technology |
| CN103226569A (en) * | 2013-03-21 | 2013-07-31 | 天脉聚源(北京)传媒科技有限公司 | Video providing method, device and system |
| CN103412870A (en) * | 2013-07-09 | 2013-11-27 | 北京深思洛克软件技术股份有限公司 | News pushing method of mobile terminal device news client side software |
| CN104036038A (en) * | 2014-06-30 | 2014-09-10 | 北京奇虎科技有限公司 | News recommendation method and system |
| CN104182549A (en) * | 2014-09-15 | 2014-12-03 | 中国联合网络通信集团有限公司 | E-mail digest generation method and device |
-
2015
- 2015-01-21 CN CN201510030145.0A patent/CN104573054B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070174167A1 (en) * | 2005-05-20 | 2007-07-26 | Stefano Natella | Derivative relationship news event reporting |
| CN101174273A (en) * | 2007-12-04 | 2008-05-07 | 清华大学 | News event detecting method based on metadata analysis |
| CN102073631A (en) * | 2009-11-19 | 2011-05-25 | 凌坚 | Video news unit dividing method by using association rule technology |
| CN103226569A (en) * | 2013-03-21 | 2013-07-31 | 天脉聚源(北京)传媒科技有限公司 | Video providing method, device and system |
| CN103412870A (en) * | 2013-07-09 | 2013-11-27 | 北京深思洛克软件技术股份有限公司 | News pushing method of mobile terminal device news client side software |
| CN104036038A (en) * | 2014-06-30 | 2014-09-10 | 北京奇虎科技有限公司 | News recommendation method and system |
| CN104182549A (en) * | 2014-09-15 | 2014-12-03 | 中国联合网络通信集团有限公司 | E-mail digest generation method and device |
Non-Patent Citations (1)
| Title |
|---|
| 陈华月 等: "基于加权关联规则的用户关注项目推荐算法", 《计算机工程》 * |
Cited By (49)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105007528A (en) * | 2015-07-06 | 2015-10-28 | 无锡天脉聚源传媒科技有限公司 | Method and device for searching video |
| CN106557508A (en) * | 2015-09-28 | 2017-04-05 | 北京神州泰岳软件股份有限公司 | A kind of text key word extracting method and device |
| CN106557513A (en) * | 2015-09-29 | 2017-04-05 | 腾讯科技（深圳）有限公司 | Event information method for pushing and event information pusher |
| CN106557513B (en) * | 2015-09-29 | 2022-05-10 | 腾讯科技（深圳）有限公司 | Event information pushing method and event information pushing device |
| US10834218B2 (en) | 2015-09-29 | 2020-11-10 | Tencent Technology (Shenzhen) Company Limited | Event information system classifying messages using machine learning classification model and pushing selected message to user |
| CN106933878A (en) * | 2015-12-30 | 2017-07-07 | 腾讯科技（北京）有限公司 | A kind of information processing method and device |
| CN106933878B (en) * | 2015-12-30 | 2021-02-05 | 腾讯科技（北京）有限公司 | Information processing method and device |
| CN106570144A (en) * | 2016-02-05 | 2017-04-19 | 中科鼎富（北京）科技发展有限公司 | Method and apparatus for recommending information |
| CN105787049B (en) * | 2016-02-26 | 2019-07-16 | 浙江大学 | A kind of network video focus incident discovery method based on Multi-source Information Fusion analysis |
| CN105787049A (en) * | 2016-02-26 | 2016-07-20 | 浙江大学 | Network video hotspot event finding method based on multi-source information fusion analysis |
| CN107402925B (en) * | 2016-05-19 | 2020-11-20 | 创新先进技术有限公司 | Information pushing method and device |
| CN107402925A (en) * | 2016-05-19 | 2017-11-28 | 阿里巴巴集团控股有限公司 | Information-pushing method and device |
| CN106202501A (en) * | 2016-07-20 | 2016-12-07 | 宁波公众信息产业有限公司 | A kind of information analysis system |
| CN106202563A (en) * | 2016-08-02 | 2016-12-07 | 西南石油大学 | A kind of real time correlation evental news recommends method and system |
| CN106372113A (en) * | 2016-08-22 | 2017-02-01 | 上海亿账通互联网科技有限公司 | News content pushing method and system |
| CN106372113B (en) * | 2016-08-22 | 2018-03-20 | 上海壹账通金融科技有限公司 | The method for pushing and system of news content |
| CN106407361A (en) * | 2016-09-07 | 2017-02-15 | 北京百度网讯科技有限公司 | Method and device for pushing information based on artificial intelligence |
| CN108241631B (en) * | 2016-12-23 | 2022-09-30 | 百度在线网络技术（北京）有限公司 | Method and device for pushing information |
| CN108241631A (en) * | 2016-12-23 | 2018-07-03 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of pushed information |
| CN106951435A (en) * | 2017-02-08 | 2017-07-14 | 广州神马移动信息科技有限公司 | News recommends method, equipment and programmable device |
| CN106993030A (en) * | 2017-03-22 | 2017-07-28 | 北京百度网讯科技有限公司 | Information-pushing method and device based on artificial intelligence |
| CN107169012A (en) * | 2017-03-31 | 2017-09-15 | 百度在线网络技术（北京）有限公司 | POI recommends method, device, equipment and computer-readable recording medium |
| CN107169012B (en) * | 2017-03-31 | 2021-03-19 | 百度在线网络技术（北京）有限公司 | POI recommendation method, device, equipment and computer readable storage medium |
| CN108694211A (en) * | 2017-04-11 | 2018-10-23 | 腾讯科技（深圳）有限公司 | Using distribution method and device |
| CN108694211B (en) * | 2017-04-11 | 2023-05-12 | 腾讯科技（深圳）有限公司 | Application distribution method and device |
| CN107239537A (en) * | 2017-06-01 | 2017-10-10 | 福建中金在线信息科技有限公司 | A kind of method of document display, device, electronic equipment and readable storage medium storing program for executing |
| CN107273355B (en) * | 2017-06-12 | 2020-07-14 | 大连理工大学 | Chinese word vector generation method based on word and phrase joint training |
| CN107273355A (en) * | 2017-06-12 | 2017-10-20 | 大连理工大学 | A kind of Chinese word vector generation method based on words joint training |
| CN107506367B (en) * | 2017-07-03 | 2021-12-24 | 创新先进技术有限公司 | Method and device for determining application display content and server |
| CN107506367A (en) * | 2017-07-03 | 2017-12-22 | 阿里巴巴集团控股有限公司 | It is determined that the method, apparatus and server of application displaying content |
| CN107657067A (en) * | 2017-11-14 | 2018-02-02 | 国网山东省电力公司电力科学研究院 | A kind of quick method for pushing of frontier science and technology information and system based on COS distance |
| CN107657067B (en) * | 2017-11-14 | 2021-03-19 | 国网山东省电力公司电力科学研究院 | Cosine distance-based leading-edge scientific and technological information rapid pushing method and system |
| CN108519980A (en) * | 2018-01-31 | 2018-09-11 | 广东易联创富集团有限公司 | News push method, apparatus, platform, computer readable storage medium |
| CN108470046A (en) * | 2018-03-07 | 2018-08-31 | 中国科学院自动化研究所 | Media event sort method and system based on media event search statement |
| CN108470046B (en) * | 2018-03-07 | 2020-12-01 | 中国科学院自动化研究所 | News event sequencing method and system based on news event search sentence |
| CN108563630A (en) * | 2018-03-21 | 2018-09-21 | 上海蔚界信息科技有限公司 | A kind of construction method of text analyzing knowledge base |
| CN109241277A (en) * | 2018-07-18 | 2019-01-18 | 北京航天云路有限公司 | The method and system of text vector weighting based on news keyword |
| CN109241277B (en) * | 2018-07-18 | 2022-05-17 | 北京航天云路有限公司 | Text vector weighting method and system based on news keywords |
| CN110110303A (en) * | 2019-03-28 | 2019-08-09 | 苏州八叉树智能科技有限公司 | Newsletter archive generation method, device, electronic equipment and computer-readable medium |
| CN110335654A (en) * | 2019-07-03 | 2019-10-15 | 重庆邮电大学 | A kind of information extraction method of electronic health record, system and computer equipment |
| CN110837609A (en) * | 2019-11-12 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Social information processing method and related product |
| CN110990690A (en) * | 2019-11-12 | 2020-04-10 | 上海易点时空网络有限公司 | Post recommendation method and device |
| CN111460289A (en) * | 2020-03-27 | 2020-07-28 | 北京百度网讯科技有限公司 | News information pushing method and device |
| CN111460289B (en) * | 2020-03-27 | 2024-03-29 | 北京百度网讯科技有限公司 | News information pushing method and device |
| CN111966784A (en) * | 2020-07-14 | 2020-11-20 | 深圳中泓在线股份有限公司 | Information recommendation method, electronic device and storage medium |
| CN112613317A (en) * | 2020-12-30 | 2021-04-06 | 中国农业银行股份有限公司 | Text data cleaning method and device |
| CN112613317B (en) * | 2020-12-30 | 2023-12-08 | 中国农业银行股份有限公司 | Text data cleaning method and device |
| CN114564675B (en) * | 2022-04-28 | 2022-07-22 | 深圳格隆汇信息科技有限公司 | Information recommendation method and device and storage medium |
| CN114564675A (en) * | 2022-04-28 | 2022-05-31 | 深圳格隆汇信息科技有限公司 | Information recommendation method and device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104573054B (en) | 2018-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104573054A (en) | Information pushing method and equipment | |
| US10621183B1 (en) | Method and system of an opinion search engine with an application programming interface for providing an opinion web portal | |
| Hoffart et al. | Discovering emerging entities with ambiguous names | |
| WO2017020451A1 (en) | Information push method and device | |
| US10002371B1 (en) | System, method, and computer program product for searching summaries of online reviews of products | |
| US20190318407A1 (en) | Method for product search using the user-weighted, attribute-based, sort-ordering and system thereof | |
| US11783132B2 (en) | Technologies for dynamically creating representations for regulations | |
| Nguyen et al. | Real-time event detection using recurrent neural network in social sensors | |
| US20140012840A1 (en) | Generating search results | |
| US10002187B2 (en) | Method and system for performing topic creation for social data | |
| CN108520007B (en) | Web page information extracting method, storage medium and computer equipment | |
| TWI571756B (en) | Methods and systems for analyzing reading log and documents corresponding thereof | |
| CN111125528A (en) | Information recommendation method and device | |
| CN112528007A (en) | Confirmation method and confirmation device for target enterprise of business inviting project | |
| CN113688310A (en) | Content recommendation method, device, equipment and storage medium | |
| CN112307336A (en) | Hotspot information mining and previewing method and device, computer equipment and storage medium | |
| CN111737607B (en) | Data processing method, device, electronic equipment and storage medium | |
| CN116089732B (en) | User preference identification method and system based on advertisement click data | |
| CN107291686B (en) | Method and system for identifying emotion identification | |
| CN114780712B (en) | News thematic generation method and device based on quality evaluation | |
| Johari et al. | The Hybrid Recommender System of the Indonesian Online Market Products using IMDb weight rating and TF-IDF | |
| CN106294689A (en) | A kind of method and apparatus selecting based on text category feature to carry out dimensionality reduction | |
| CN107622125B (en) | Information crawling method and device and electronic equipment | |
| CN110147488B (en) | Page content processing method, processing device, computing equipment and storage medium | |
| KR102041915B1 (en) | Database module using artificial intelligence, economic data providing system and method using the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 310052 Room 301, Building No. 599, Changhe Street Network Business Road, Binjiang District, Hangzhou City, Zhejiang Province Patentee after: Hangzhou NetEase Shuzhifan Technology Co.,Ltd. Address before: 310052 Room 301, Building No. 599, Changhe Street Network Business Road, Binjiang District, Hangzhou City, Zhejiang Province Patentee before: HANGZHOU LANGHE TECHNOLOGY Ltd. |